April 25, 2011
File naming best practices for digital asset management
What’s in a filename?
Chances are that you use a desktop computer and already know that filenames are used for storing, organizing, and retrieving computer files. Every time you open an existing file, or save a new file, you’re given an opportunity to enter a descriptive and useful filename that will come in handy later on. Sometimes, however, we don’t spend enough time thinking about what goes into that “Save As” box or stick with the default “Untitled.xxx” or “First few words of your document.doc” filename. Multiply this scenario times the number of users on a fileserver, and you soon run into problems.
If only there was a system that would transform disorganized messes of filenames into easily searchable databases making filenames obsolete! Did you think I was going to say digital asset management is such a solution? Well, kind of. Read on…
Filenames are so 2004
With a digital asset management system, filenames may seem less important since most DAMS do not require unique filenames. DAMS provide search features that allow users to locate files based on metadata like keywords, descriptions or custom fields. Often this metadata is already embedded in the file and available automatically to the DAM, or is applied by the person adding files to the DAM. Hypothetically you could have 100,000 “untitled.jpg” files in your DAM with accurate metadata and everything would work great. Problems with duplicate filename collisions where a filename already exists in a folder are often avoided by operating systems and DAMS that append the filename to prevent a conflict. Many workflows do not rely on filenames for these reasons, but let’s take a look at a few reasons why filenames sometimes matter with DAM.
Keywords from path
DAM systems often include features for automatically creating keyword metadata from the file’s path, which includes filename and folder names. With this feature, well thought out filenames can be a huge timesaver for anyone adding or searching the DAM since this metadata is generated automatically. A file saved as “Part 032423.jpg” automatically shows up in a search for “032423”.
The Original Metadata
Filenames are the original metadata, or “data about data”, for the file. Before we had fancy EXIF Date Taken, Photoshop Rating, and embedded XMP keyword metadata we relied on filenames to help us find and figure out what was in files. Every program on your computer that can open and save files “supports” filename metadata, so this metadata works almost everywhere! No fancy tools or applications are needed to view the filename – just look at the file.
So you have all this fancy, organized, detailed metadata in your DAM. Maybe you’re even embedding the metadata back into the file for other applications to read and process. What happens when you send the file to someone else outside of the DAM? Does the metadata stay with the file? Do their systems support reading metadata? Do they know that the file has metadata? Do they even know how to view the metadata? It’s almost certain that the filename will still be there on the other end, right in front of their face, providing information about what’s inside.
In many systems that display lists of files, you can set the sort order based on filenames or other attributes. When sorting on filenames, the file naming convention will affect the order files are displayed. This can be a problem when a MONTH-DAY-YEAR date format is used in filenames since the sort will group first by month, then by day, followed by year. To chronologically sort files based on date, use a YEAR-MONTH-DAY format along with a 4 digit year as opposed to a 2 digit year. For example, 20110413.jpg instead of 04132011.jpeg.
Continuing the example of distributing files to someone else, what happens when they have a question about what you sent them? If they ask about “DSC0010.jpg”, does that mean anything or do you have hundreds of files with that name? Incorporating a file naming convention that results in unique filenames can provide a key (or unique identifier) that links back to one and only one file. Another useful advantage with unique filenames is the ability to import and synchronize data with other systems that key off unique filenames. When cataloging files in a multi-user environment, one simple file naming convention to ensure unique and chronological filenames is using the date, the cataloging “session” (if the user catalogs more than once a day), and the cataloger’s username followed by a number that increments per file. For example, 20110413_a_esmith_001.jpg.
Cross platform limitations
Filenames can be problematic when sharing files between platforms and systems. For example; Mac allows certain symbols in filenames that Windows does not. For files on the web, spaces are replaced with %20 in URLs turning “A filename you can easily read.doc” into “A%20filename%20you%20can%20easily%20read.doc”. In general, try to stay away from spaces in filenames as well as the following characters:
/ : * ? “ < > | [ ] & $
These scenarios don’t apply to everyone, and many successful workflows use DAM to leverage metadata and “cover up” their mess of disorganized filenames. An excellent way to address the issues above is to choose a file naming convention for adding files to your DAM, and batch rename files before or after they are added to the DAM. A batch rename tool or feature can apply your file naming convention across many files in a single operation as opposed to renaming each file manually.
Do you care about filenames? What do you think is important in a filename? Let us know in the comments section below.